A distributed processing system is known in which, when a master node receives a data access request from any one of a plurality of nodes (computers), the master node inquires of each node whether data is stored in a page cache, and responds, to a request source node requesting an access, with information specifying a node responding that data is to be stored in a page cache. (PTL 1)